cs.CL（2026-08-26）

📊 共 25 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (20 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (3 🔗1) 支柱四：生成式动作 (Generative Motion) (1) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (20 篇)

#题目一句话要点标签🔗⭐
1 OmniPhys: A Unified Multimodal Benchmark for Physics Understanding and Generation from Chinese Educational Corpora 提出OmniPhys以解决物理领域多模态基准缺乏问题 large language model multimodal ✅
2 PlanSightRAG: A Visual-First Multimodal RAG for Automating Question Answering and Compliance Checking for Civil Standard Plans 提出PlanSightRAG以解决土木标准图纸合规检查问题 multimodal
3 Think-Probe-Respond: Improving Large Language Models as Judges of Research Idea Novelty 提出Think-Probe-Respond以解决大语言模型判断研究创意新颖性的问题 large language model
4 Distinct dynamics of conceptual and referential disruptions in human reading and large language model processing 研究概念与指称信息在阅读中的动态差异 large language model
5 Generative vs. Encoder Large Language Models for ASR Evaluation: A Comparative Study 比较生成与编码大型语言模型在ASR评估中的应用 large language model
6 Anchoring Bias in LLM-as-a-Judge Systems: Prior Scores Compromise Evaluation Independence 揭示大语言模型评估中的锚定偏差问题 large language model chain-of-thought
7 Skill Issue: Are Skills Language-Invariant in LLMs? 量化语言模型跨语言技能不一致性以提升多语言性能 large language model
8 JIT-Agent: Scaling Harness Intelligence via Just-in-Time Harness Evolution 提出JIT-Agent以解决智能体工具设计的可扩展性问题 foundation model
9 Beam Search, Self-Consistency, and the Limits of Inference-Time Scaling for Grammar-Constrained Text-to-SQL in Small Language Models 提出基于束搜索和自一致性的文本到SQL转换方法以优化小型语言模型 large language model
10 Controllable Affective Generation via Latent Vector Steering 提出EmoVec以解决大语言模型情感生成不足问题 large language model
11 Adaptive Triggering for Bias Correction in LLM Reasoning 提出自适应触发机制以解决LLM推理中的偏见问题 chain-of-thought
12 A Self-Evolving Multi-Agent Framework Defense against LLM Jailbreak Attacks 提出自演化多智能体框架以防御LLM越狱攻击 large language model
13 Unveiling Spectral Mechanisms in Training-Free LLM Text Detection 提出频谱分析方法以解决训练无关的文本检测问题 large language model
14 From Passive Response to Proactive Correction: Enhancing LLM Robustness Against Input Fact Perturbations 提出DEDUCE框架以增强LLM对输入事实扰动的鲁棒性 large language model
15 Localize-Then-Decide Guarantees for LLM Judgments 提出Localize-Then-Decide框架以解决LLM判断一致性问题 large language model
16 GRIP: Granular Reward-Guided Parameter Interpolation for Efficient Reasoning 提出GRIP框架以实现高效推理的参数插值 large language model
17 Cross-Dataset Stability of Expert-Informed Skill Prompting and Fine-Tuning for Chinese Metaphor Identification 提出专家指导的技能提示以提升中文隐喻识别的跨数据集稳定性 large language model
18 VietAIDetector: An Open-Source Zero-Shot Detector for Vietnamese AI-Generated Text 提出VietAIDetector以解决越南语AI生成文本检测问题 large language model ✅
19 Provenance Before Prose: Claim-Locked Reporting 提出Claim-Locked Reporting以解决统计报告的可重复性问题 large language model
20 Groundhog Bit-Flip Attack: Seeding Infinite Generation Loops in Mixture-of-Experts LLMs through Bit Flips 提出Groundhog Bit-Flip攻击以揭示MoE LLM的脆弱性 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
21 VISA: Agentic Self-Evolving Data Synthesis for Multimodal Instruction Following 提出VISA框架以解决多模态指令跟随数据合成问题 reinforcement learning large language model multimodal
22 MoganBert-TR: A Turkish Encoder Foundation Model Trained from Scratch with a CLM-to-MLM Curriculum 提出MoganBert-TR以提升土耳其语编码模型性能 distillation foundation model ✅
23 From Memorization to Absorption: Mixed-Policy RL for Continual Knowledge Injection 提出GRIN框架以解决持续知识注入问题 reinforcement learning large language model

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
24 DCGC: Draft-Conditioned Global Correction for Complex Reasoning with Masked Diffusion Models 提出DCGC以解决大型语言模型推理错误问题 MDM large language model

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
25 EgoArgus: Benchmarking VLMs as Situational Assistants for Modality-Grounded User Supports 提出EgoArgus以解决VLMs在日常对话中的决策问题 egocentric

⬅️ 返回 cs.CL 首页 · 🏠 返回主页